FILED SENATE
Apr 6, 2021
GENERAL ASSEMBLY OF NORTH CAROLINA
S.B. 581
SESSION 2021 PRINCIPAL CLERK
S D
SENATE BILL DRS15230-MT-87
Short Title: Redistricting Criteria for 2021. (Public)
Sponsors: Senators Clark and Fitch (Primary Sponsors).
Referred to:
1 A BILL TO BE ENTITLED
2 AN ACT TO ESTABLISH CRITERIA FOR LEGISLATIVE AND CONGRESSIONAL
3 REDISTRICTING FOLLOWING THE RETURN OF THE 2020 DECENNIAL CENSUS.
4 Whereas, following the receipt on March 2, 2011, of population data from the 2010
5 decennial census pursuant to P.L. 94-171 (2010 Redistricting Data File), the General Assembly
6 realigned districts for the following bodies on the following dates:
7 (1) House of Representatives of the United States Congress on July 28, 2011, in
8 S.L. 2011-403, as amended by S.L. 2011-414, hereinafter referred to as Senate
9 Bill 453.
10 (2) North Carolina Senate on July 27, 2011, in S.L. 2011-402, as amended by S.L.
11 2011-413, hereinafter referred to as Senate Bill 455.
12 (3) North Carolina House of Representatives on July 28, 2011, in S.L. 2011-404,
13 as amended by S.L. 2011-416, hereinafter referred to as House Bill 937; and
14 Whereas, on February 5, 2016, the United States District Court for the Middle District
15 of North Carolina held in Harris v. McCrory, 159 F. Supp. 3d 600, that Senate Bill 453 was an
16 unconstitutional racial gerrymander; and
17 Whereas, on February 19, 2016, the General Assembly enacted a remedial plan for
18 congressional districts in S.L. 2016-1, hereinafter referred to as Senate Bill 2; and
19 Whereas, on October 28, 2019, a three-judge panel of the superior court division of
20 the General Court of Justice in Harper v. Lewis, 19 CVS 012667, concluded that the
21 congressional districts enacted in Senate Bill 2 were unconstitutional extreme partisan
22 gerrymanders and enjoined the State from holding elections under those districts; and
23 Whereas, on November 15, 2019, the General Assembly enacted a remedial plan for
24 congressional districts for the 2020 general election in S.L. 2019-249, hereinafter referred to as
25 House Bill 1029; and
26 Whereas, on August 11, 2016, the United States District Court for the Middle District
27 of North Carolina held in Covington v. North Carolina, 316 F.R.D. 117, aff'd, 137 S. Ct. 2211,
28 that portions of Senate Bill 455 and House Bill 937 were unconstitutional racial gerrymanders;
29 and
30 Whereas, on August 31, 2017, the General Assembly enacted remedial plans for
31 legislative districts for use beginning with the 2018 general election in S.L. 2017-208, hereinafter
32 referred to as House Bill 927, and S.L. 2017-207, hereinafter referred to as Senate Bill 691; and
33 Whereas, on January 21, 2018, the United States District Court for the Middle District
34 of North Carolina held in Covington v. North Carolina, 283 F. Supp. 3d 410, aff'd in part and
35 rev'd in part, 138 S. Ct. 2548, that certain districts realigned in House Bill 927 and Senate Bill
*DRS15230-MT-87*
 General Assembly Of North Carolina Session 2021
1 691 continued to be unconstitutional racial gerrymanders and instituted its own remedial districts
2 for use beginning with the 2018 general election; and
3 Whereas, on November 2, 2018, a three-judge panel of the superior court division of
4 the General Court of Justice in NAACP v. Lewis, 18 CVS 002322, held that certain districts
5 realigned by the General Assembly in 2017 violated the North Carolina Constitution's prohibition
6 against mid-decade redistricting; and
7 Whereas, on September 3, 2019, a three-judge panel of the superior court division of
8 the General Court of Justice in Common Cause v. Lewis, 18 CVS 014001, held that additional
9 portions of House Bill 927 and Senate Bill 691 were unconstitutional partisan gerrymanders; and
10 Whereas, on September 17, 2019, the General Assembly enacted remedial plans for
11 legislative districts for use in the 2020 general election in S.L. 2019-220, hereinafter referred to
12 as House Bill 1020, and S.L. 2019-219, hereinafter referred to as Senate Bill 692; and
13 Whereas, on October 28, 2019, the three-judge panel of the superior court division of
14 the General Court of Justice approved the remedial maps for use in the 2020 general election;
15 and
16 Whereas, every congressional and legislative election conducted in the State of North
17 Carolina during the 2010 decade was conducted with the use of unconstitutional congressional
18 and legislating districting plans that contained either racial gerrymanders, partisan gerrymanders,
19 or both; and
20 Whereas, it is the intent of the General Assembly to avoid racial and partisan
21 gerrymanders in future congressional and legislative districts; Now, therefore,
22 The General Assembly of North Carolina enacts:
23 SECTION 1. Following the return of the 2020 federal decennial census, for the
24 purpose of revising districts and the apportionment among those districts of members of the
25 Senate and the House of Representatives of the General Assembly and the House of
26 Representatives of the United States Congress, the following requirements shall apply:
27 (1) Baseline criteria. – Baseline criteria, as defined below, shall have priority over
28 any other redistricting criteria. For purposes of this act, baseline criteria refers
29 to all of the following, in order of priority:
30 a. Equal population. – Each member of each body identified above shall
31 represent, as nearly as may be, an equal number of inhabitants. The
32 ideal population for a district is the population of the State, as reported
33 by the 2020 federal decennial census, divided by the number of
34 members in a plan for one of the bodies identified above.
35 b. Population deviation. – For purposes of this act, "total population
36 deviation" refers to the difference between the population of the most
37 populous district and the least populous district, and "population
38 deviation from ideal" refers to the difference between the actual
39 population of a district and the ideal population for that district.
40 Population deviations for each body identified above shall be as
41 follows:
42 1. Congress. – Population deviation from ideal shall be zero or
43 one person, unless a higher deviation is necessary to achieve
44 or optimize a compelling State interest associated with the
45 baseline criteria.
46 2. North Carolina Senate and House of Representatives. – Total
47 population deviation shall not exceed ten percent (10%).
48 Population deviation from ideal shall not exceed five percent
49 (5%), in accordance with Stephenson v. Bartlett, 355 N.C. 354,
50 562 S.E.2d 377 (2002).
Page 2 DRS15230-MT-87
 General Assembly Of North Carolina Session 2021
1 c. Contiguity. – All districts shall be contiguous. Contiguity by water is
2 sufficient. To the extent practicable, areas within a district should be
3 easily accessible to one another without requiring travel through
4 another district.
5 d. County groupings. – Legislative districts shall be drawn within county
6 groups as required by Stephenson v. Bartlett, 355 N.C. 354, 562 S.E.2d
7 377 (2002), Stephenson v. Bartlett, 357 N.C. 301, 582 S.E.2d 247
8 (2003), Dickson v. Rucho, 367 N.C. 542, 766 S.E.2d 238 (2014), and
9 Dickson v. Rucho, 368 N.C. 481, 781 S.E.2d 460 (2015). Within
10 county groupings, county lines shall not be crossed except as
11 authorized by the cases identified in this sub-subdivision.
12 e. Political boundaries. – All districts shall minimize the number of split
13 precincts and municipalities.
14 f. Communities of interest. – All districts shall minimize the number of
15 split communities of interest. For purposes of this act, "communities
16 of interest" are geographically contiguous areas of cohesive
17 populations of people that share common social, cultural, and
18 economic interests that should be included within a single district for
19 purposes of their effective, fair, and equitable representation. A
20 community of interest does not include a community based on political
21 affiliation or relationships with a political party, elected official, or
22 candidate for office. Public and private institutions of higher education
23 that offer a postsecondary degree, as defined in G.S. 116-15(a2)(1),
24 and have a residential campus, including off-site housing near the
25 campus, constitute communities of interest.
26 g. Compactness. – Reasonable efforts shall be made to ensure that all
27 districts are compact. The following measures shall be used for
28 assessing compactness:
29 1. The number of cut edges in a plan, as described in
30 Recombination, A family of Markov chains for redistricting by
31 Daryl DeFord, Moon Duchin, and Justin Solomon in an article
32 published on March 27, 2020, and available at
33 https://mggg.org/uploads/ReCom.pdf.
34 2. Reock, i.e., dispersion, and Polsby-Popper, i.e., perimeter,
35 assessments.
36 (2) Candidate considerations. – No effort shall be made to create a district
37 favorable or unfavorable to any candidate.
38 (3) Partisan advantage. – No effort shall be made to maintain or establish an
39 electoral advantage for any party in any plan. Based on an outlier analysis
40 conducted in accordance with subdivision (6) of this section, except as
41 necessary to comply with State and federal law, a plan shall not advantage a
42 political party beyond the most common seat distribution for that plan, except
43 as follows:
44 a. For a congressional plan, by no more than one district.
45 b. For a plan for the North Carolina Senate, by no more than two districts.
46 c. For a plan for the North Carolina House of Representatives, by no
47 more than three districts.
48 (4) Partisan election data. – Election results data may only be used as part of an
49 ensemble analysis of an entire plan, including an outlier analysis, as provided
50 in subdivision (6) of this section. Election results data shall not be used in
51 order to provide any party a disproportionate number of seats in a plan, and a
DRS15230-MT-87 Page 3
 General Assembly Of North Carolina Session 2021
1 composite index of election results shall not be used. Only election results data
2 from elections for the following offices occurring in and after the year 2016
3 shall be considered:
4 a. All offices of the Council of State.
5 b. President of the United States.
6 c. United States Senate.
7 (5) Partisan analysis. – To add context and validity to the outlier analysis
8 performed pursuant to subdivision (6) of this section, the third-party
9 consultant identified in subdivision (6) of this section shall produce
10 rank-ordered marginal histograms that show typical vote fractions of all
11 districts in each plan from the district that favors each political party the most
12 to the district that favors each political party the least. Additionally, all plans
13 shall be evaluated based on elections from each general election in at least the
14 previous 10 years for each of the offices identified in sub-subdivisions a., b.,
15 and c. of subdivision (4) of this section with different statewide vote counts.
16 To the extent possible, the information produced pursuant to this subdivision
17 shall comport with the methodology discussed in the article Quantifying
18 Gerrymandering in North Carolina by Gregory Herschlag, Han Sung Kang,
19 Justin Luo, Christy Vaughn Graves, Sachet Bangia, Robert Ravier, and
20 Jonathan C. Mattingly, published in volume 7, issue 1, of the 2020 edition of
21 the journal Statistics and Public Policy.
22 (6) Outlier analysis. – All districting plans shall be subjected to an analysis of
23 their probable partisan effects prior to their adoption by any committee of the
24 General Assembly or enactment by the General Assembly. This process shall
25 be performed by a third-party consultant. The third-party consultant shall
26 produce at least all of the following:
27 a. An ensemble of at least 20,000 alternative plans that meet the
28 requirements of this section. Election data shall not be used in the
29 construction of the ensemble.
30 b. An analysis of the plans using a method for which the consultant shall
31 provide a detailed description.
32 c. Evidence that the number of plans drawn for the analysis is sufficient
33 for the statistics and diagrams presented to have stabilized.
34 d. Evidence that choices made in generating the plans are consistent with
35 the policy priorities specified in this section and do not affect
36 qualitative outcomes.
37 (7) Summary metrics. – The following summary metrics shall be used as part of
38 the outlier analysis described in subdivision (6) of this section:
39 a. Declination. – The method developed by Gregory S. Warrington to
40 identify possible partisan gerrymanders by analyzing voter
41 distributions.
42 b. Gerrymandering index. – The method developed by Jonathan
43 Mattingly to quantify and provide relative context for packing and
44 cracking in districting plans by measuring how individual districts
45 deviate from an expected percentage of partisan voters.
46 (8) Consultant disclosure. – Notwithstanding any other provision of law, if any
47 member, committee, officer, or employee of the General Assembly hires or
48 consults with any person or entity not employed by the General Assembly
49 regarding the realignment of districts for any plan, all related information is
50 no longer confidential and is a public record. The member, committee, officer,
51 or employee of the General Assembly shall publish the name of the person or
Page 4 DRS15230-MT-87
 General Assembly Of North Carolina Session 2021
1 entity and all communications with that person or entity within 24 hours of
2 hiring that person or entity and receiving any communication from that person
3 or entity.
4 (9) Map source disclosure. – If any member, committee, officer, or employee of
5 the General Assembly receives a plan to realign districts from any person or
6 entity that is not a member of or employed by the General Assembly, the
7 member, committee, officer, or employee shall publish the plan and the name<